doocus update and meeting list export

tests/test_doocus.py

@@ -13,6 +13,7 @@ from pathlib import Path
 from tests import REPO_ROOT  # noqa: F401  (puts repo root on sys.path)
 from doocus import registry
 from doocus.workflow import DocConfig, DocWorkflow
+from doocus.tree import build_index, classify
 
 
 def has(mod: str) -> bool:
@@ -136,5 +137,59 @@ class TestExtractors(unittest.TestCase):
         self.assertEqual(meta["title"], "PDF Doc")
 
 
+class TestTreeIndex(unittest.TestCase):
+    def test_classify(self):
+        self.assertEqual(classify("docx"), "extracted")
+        self.assertEqual(classify("pdf"), "extracted")
+        self.assertEqual(classify("mp4"), "meeting")
+        self.assertEqual(classify("mkv"), "meeting")
+        self.assertEqual(classify("md"), "link")
+        self.assertEqual(classify("csv"), "link")
+        self.assertEqual(classify("bin"), "link")
+
+    def test_build_index_replicates_tree(self):
+        with tempfile.TemporaryDirectory() as tmp:
+            src = Path(tmp) / "drive"
+            (src / "team a" / "sub").mkdir(parents=True)
+            (src / "team a" / "notes.md").write_text("# Hi\n\nbody\n")
+            (src / "team a" / "sub" / "t.csv").write_text("a,b\n1,2\n")
+            (src / "team a" / "clip.mp4").write_bytes(b"\x00\x00")   # video → meeting
+            (src / "misc.bin").write_bytes(b"\x00")                 # unknown → link
+            out = Path(tmp) / "out"
+
+            index = build_index(src, out)
+
+            # Whole tree replicated (no flattening), paths preserved.
+            paths = {f["path"] for f in index["files"]}
+            self.assertEqual(paths, {
+                "team a/notes.md", "team a/sub/t.csv", "team a/clip.mp4", "misc.bin"})
+            modes = {f["path"]: f["mode"] for f in index["files"]}
+            self.assertEqual(modes["team a/clip.mp4"], "meeting")
+            self.assertEqual(modes["team a/notes.md"], "link")
+            self.assertEqual(modes["misc.bin"], "link")
+            # Every node carries the Drive-url slot.
+            self.assertTrue(all("url" in f for f in index["files"]))
+            self.assertTrue((out / "index.json").exists())
+
+    @unittest.skipUnless(has("docx"), "python-docx not installed")
+    def test_build_index_extracts_sidecar(self):
+        import docx
+        with tempfile.TemporaryDirectory() as tmp:
+            src = Path(tmp) / "drive"
+            src.mkdir()
+            doc = docx.Document()
+            doc.add_paragraph("hello world")
+            doc.save(src / "report.docx")
+            out = Path(tmp) / "out"
+
+            index = build_index(src, out)
+            node = next(f for f in index["files"] if f["path"] == "report.docx")
+            self.assertEqual(node["mode"], "extracted")
+            sidecar = out / node["out"]
+            self.assertTrue((sidecar / "content.md").exists())
+            self.assertTrue((sidecar / "meta.json").exists())
+            self.assertIn("hello", (sidecar / "content.md").read_text())
+
+
 if __name__ == "__main__":
     unittest.main()